Hello, my configuration is the following:
1 server with Kamailio:5060, CDR Master, Routing Master, Redis Master, Sems:5061
1 server with CDR Slave, Routing Slave, Redis slave, Sems:5060
I’ve configured redis master with this parameter:
notify-keyspace-events “Egx”
bind 127.0.0.1 MYMAINIP
Redis slave i’ve added only
bind 127.0.0.1
replicaof MYMAINIP 6379
On sems.conf of both servers I put 127.0.0.1 as redis ip
But on the slave server I receive these errors, basically he tries to write something to slave redis I think:
Mar 30 10:59:53 Debian-bullseye-latest-amd64-base sems[209964]: [210028/yeti:resources/ResourceControl.cpp:193] ERROR: ResourceControl::on_reconnect(): invalidate resources
Mar 30 10:59:53 Debian-bullseye-latest-amd64-base sems[209964]: [210028/yeti:resources/ResourceCache.cpp:380] ERROR: ReplyTypeException MULTI HSET type not desired type: 6
Mar 30 10:59:53 Debian-bullseye-latest-amd64-base sems[209964]: [210028/yeti:alarms.cpp:56] ERROR: ALARM 5 [redis write connection error] RAISED. current: 1
Mar 30 10:59:53 Debian-bullseye-latest-amd64-base sems[209964]: [210028/yeti:resources/ResourceControl.cpp:195] ERROR: can’t clear resources. please examine actual state. (all handlers invalidated)
Mar 30 10:59:53 Debian-bullseye-latest-amd64-base sems[209964]: [210028/yeti:resources/ResourceControl.cpp:193] ERROR: ResourceControl::on_reconnect(): invalidate resources
Mar 30 10:59:53 Debian-bullseye-latest-amd64-base sems[209964]: [210028/yeti:resources/ResourceCache.cpp:380] ERROR: ReplyTypeException MULTI HSET type not desired type: 6
Mar 30 10:59:53 Debian-bullseye-latest-amd64-base sems[209964]: [210028/yeti:alarms.cpp:56] ERROR: ALARM 5 [redis write connection error] RAISED. current: 1
How can I fix it?
Do I have to put on sems.conf the remote address of the master redis?
This issue is very resource intensive, consider that on master server I have a total load of 3% while on slave server it peaks at 45% because of redis-server process hanging and systemd-journal updating every second.
Thank you